526 HISTORY OF SCIENCE. 



atoms have a power latent in them a mechanical power ready for 

 action like that of a wound-up watch-spring. 



In 1845 Mr. Joule estimated the heat developed, when air is com- 

 pressed. The apparatus employed was very simple in principle, con- 

 sisting merely of a strong copper vessel and a pump for forcing air 

 into it up to a pressure of more than 20 atmospheres. The pump 

 and receiver were both immersed in a large vessel of water, and a very 

 delicate thermometer was employed in ascertaining the temperatures 

 of the water. It will be unnecessary to detail here all subsidiary pre- 

 cautions and corrections employed in the conduct of the experiments, 

 as, for instance, how the air was supplied to the pump dry and of 

 uniform temperature ; how the effects of friction were eliminated, etc. 

 The results gave 795 foot-lbs. as the mechanical equivalent of the 

 heat of a pound of water raised one degree. A very ingenious ex- 

 periment was devised to prove that the mere change in the position 

 of the particles of the air their greater or less approximation does 

 not of itself involve any development or absorption of heat. Two 

 copper receivers, v, c, Fig. 243, were provided, 

 which could be connected together by a pipe 

 provided with a stop-cock, #, of a construction 

 suitable for high pressures. Into the receiver 

 c dry air was condensed to about 22 atmo- 

 spheres, and on the other hand v was completely 

 exhausted. The two receivers were then con- 

 nected by a coupling-piece, b, and placed in a 

 FIG. 243. tin vessel containing water. The temperature of 



the water was ascertained, and then communica- 

 tion was opened between the two vessels. The air of course rushed 

 from c into v until the pressures within the two vessels had become 

 equal. The temperature of the water was again noted and found to 

 have undergone no change. The experiment was modified by placing 

 the charged vessel, c, in one pan, and the vacuous vessel, v, in another. 

 When equilibrium had been restored by opening the stop-cocks, it 

 was found that the temperature of the water surrounding the vessel 

 from which the air had expanded had fallen 2 0> 38, while that sur- 

 rounding the other receiver had gained 2 '3 6 ; that is to say, the 

 quantity of heat lost by one receiver had been acquired by the 

 other. In other words, the total heat produced or absorbed by the 

 operation was nil. The compressed air, expanding into a vacuum to 

 double its former volume, had performed no mechanical work in 

 reaching the state of equilibrium ; nor had any work been expended 

 upon it by any external agent : hence there was neither loss nor gain 

 of heat. On the other hand, when one of the receivers containing 

 condensed air was placed in a vessel of water, and communication 

 opened (under proper conditions) with the external atmosphere, the 

 fall of the temperature of the water showed a loss of heat. In this 



